The initial treatment of drug addiction in a hospital setting may proceed smoothly. However, upon return to the patient's environment. intense craving often occurs and leads to relapse. Both human and animal studies have revealed that such cravings result from the association of environmental stimuli with the state of psychoactive change in a Pavlovian manner. These conditioned stimuli are powerful enough to elicit dramatic increases in craving and activation of certain brain regions in human addicts as well as behavioral changes in animals formerly treated with drugs of abuse. To study conditioned responses that may elicit craving, we will use conditioned locomotion in rats. This refers to the ability of a stimulus alone to elicit locomotor activity after it is paired repeatedly with cocaine. A discrete cue (flashing light) will be used to enable precise investigation of the control this stimulus exerts over neurochemical changes. The hypothesis to be tested is that the expression of conditioned locomotion requires increased glutamate transmission in neural circuits involving the prefrontal cortex, basolateral amygdala and nucleus accumbens. Preliminary experiments demonstrated the acquisition of conditioned locomotion to the flashing light as well as robust conditioned responses 9 days after the last pairing with cocaine. Aim l will further characterize the time course and resistance to extinction of cocaine conditioned locomotion. Aim 2 will examine the anatomical regions necessary for expression of conditioned locomotion. We had planned to use Fos as a marker for neuronal activation but found that it was not induced by conditioning with the discrete flashing light cue. Thus, regions historically found to be relevant for conditioned drug responses will be examined through lesions and lidocaine inactivation. Using the information gained from Aim 2, Aim 3 will use microdialysis to measure glutamate levels in these regions during presentation of the conditioned cue. Determining the site and pharmacological nature of the conditioned change would provide information to guide treatment attempts to block expression of such responses and thus prevent relapse.